Beruflich Dokumente
Kultur Dokumente
E d
S = 4k
e
Q
S
,
1
See Maxwells Equations (MISN-0-146).
2
See Gausss Law and Spherically Distributed Charges (MISN-0-132).
B
`
C
`
d
`
l
Figure 1. An in-
tegration path C in
a magnetic eld B.
The associated cur-
rents are not shown.
5
MISN-0-138 2
current
in - direction
current
in + direction
Figure 2. A right-hand rule assigns a sign to each current
bounded by the loop C.
while Amperes law involves a line integral:
3
Amperes law:
_
C
B d
= 4k
m
I
C
. (1)
Here:
k
m
is the magnetic force constant.
_
C
denotes integration along a closed imaginary line C.
4
The closed
imaginary line for any particular problem is usually called the in-
tegration loop or path for that problem. The line must pass
through the point where you want to know the magnetic eld.
I
C
denotes the net electric current passing through any (imaginary)
surface whose boundary is the same closed line C used in
_
C
(see
Fig. 1).
5
d is an innitesimal element of length along the integration line.
The direction of integration around the line is arbitrary, but once
taken it xes the direction of current that must be called positive.
The relevant rule will be taken up later.
3
See the Appendix of this module for a discussion of line integrals.
4
The word closed means that the line has no end so it must be a closed loop.
5
All surfaces bounded by the same line give the same value for I
C
.
6
MISN-0-138 3
wire
I out of page
r
`
d
`
l
B
`
Figure 3.
B and d
are both
tangent to C, the circular inte-
gration path.
2b. Determining Signs (). The algebraic sign () of any current
enclosed by the integration loop in Amperes law is determined by a right-
hand rule:
A current is taken to be positive if it points in the direction of the
thumb on the right hand when the ngers of that hand encircle the
loop in the direction that the line integral is taken (see Fig. 2). If it
is in the opposite direction, the current must be taken as negative.
3. Simple Applications
3a. Magnetic Field Near a Long Thin Wire. The magnetic eld
so
_
B d
= 2B = 4k
m
nI , (7)
where n is the number of wires per unit length and n is the total number
enclosed. Solving for B gives:
B = 2k
m
nI . (8)
9
MISN-0-138 6
x x x x x x x x x
l
Figure 7. The integration path for
a very long solenoid having B = 0
outside.
This equation indicates that the eld associated with an innite plane of
current is independent of the distance from the plane.
4. Example Devices
4a. The Solenoid. A solenoid is a tightly wound cylindrical helix
of current-carrying wire, used to make an electrical signal cause a one-
directional mechanical force (for example, operating a plunger). Solenoids
are frequently encountered in science and technology; there are at least
several in every car. The magnetic eld inside a solenoid can be easily
found using Amperes law. The external magnetic eld due to a solenoid of
nite length is quite similar to that of a bar magnet (see Fig. 6). However,
if the solenoid is very long, (i.e., if its length is much greater than its
radius), the eld outside is essentially zero, and inside the solenoid it is
uniform and parallel to the solenoids axis (see Fig. 7).
8
4b. Calculating the Field of a Solenoid. The magnitude of
B inside
a solenoid can be found by applying Amperes law to the rectangular
integration path shown in Fig. 7. Outside the solenoid
B is zero. Inside,
B d
= B . (9)
8
Look at the solenoid in Fig. 6 and notice that the magnetic eld lines are much
more dense inside the solenoid than outside it. Imagine making the solenoid longer
and longer, during which the density inside remains constant but the density outside
becomes more and more sparce.
10
MISN-0-138 7
r
I
I
circle of
integration
Figure 8. A toroid.
r
`
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
C
Figure 9. The circular
path of integration in-
side a toroid.
The net current through the rectangle is nI, where n is the number of
turns per unit length over the entire length . Amperes law then gives
for the magnetic eld:
B = 4k
m
nI , (solenoid) (10)
indicating a uniform eld.
4c. The Toroid. A toroid is a solenoid that has been bent into a
circle, assuming the space-saving shape of a doughnut (see Fig. 8). The
magnetic eld inside a toroid carrying a current I can be found using
Amperes law. By symmetry, the magnetic eld is tangent to the circular
integration path shown in Fig. 9. Therefore:
_
C
B d
= (B)(2r) , (11)
and the enclosed current is NI, where N is the total number of turns on
the solendoid. Then:
B = 2k
m
_
NI
r
_
. (12)
Notice that, unlike the solenoid, the magnetic eld inside the toroid is
not constant over the cross section of the coil but varies inversely as the
distance r. For points outside a toroid, it can be shown that the eld is es-
sentially zero if the turns of wire are very close together. Help: [S-1]
11
MISN-0-138 8
j
`
ds
n
^
Figure 10. The current density
j is not always
along the normal n to an arbitrary surface ele-
ment dS.
5. Using Current Density
5a. Introduction. Just as it is often useful to use the concept of
charge density in electrostatics, in magnetics we often use the concept of
current density. Charge density is a scalar and has three varieties: linear,
surface and volume. Current density is a vector and has one variety.
Current density has a non-zero value only at those space-points where
there are charges owing so there is an electric current: it is the net
amount of charge going through the space-point per unit time, per unit
area perpendicular to the direction of the current. The direction of the
current density vector is the direction of the electric current at the space-
point in question. The universal symbol for the current density is
j(r),
where the argument indicates that the current density may change as one
moves from one space-point to another.
9
5b. The Current Through a Surface. We now assume we know
the current density
j at various space-points of interest and we want to
nd the I
C
used in Amperes law, Eq. (1). We start with
j at the point
of a surface element dS having a normal unit vector n. We want to know
how much current dI is passing through this element of surface. Since
j
is the current per unit area normal to the current, we must multiply by
an element of area dA normal to the current (see Fig. 10). If we know dS,
n
S
, and
j, we can get dA by (see Appendix B):
dA =
j ndS .
Substituting dI = jdA we get:
dI =
j ndS . (13)
For the special case of a uniform current owing perpendicular to a plane
surface of area A, the equation simplies to I = jA; a simple statement
9
Of course
j may also be a function of time but we are not dealing with that case
here.
12
MISN-0-138 9
C
a
r
b
Figure 11. A hollow conducting
cylinder with a non-uniform current
density j = k/r.
that the current is the current density times area.
Finally, we integrate both sides of Eq. (13) to get:
I
C
=
_
S
j ndS . (14)
5c. Amperes Law in Terms of the Current Density. Amperes
law may be rewritten in terms of the current density, using Eqs. (1) and
(14), giving:
_
C
B d
= 4k
m
_
S
j ndS . (15)
Here C is the closed path around the perimeter of the surface S.
5d. Example: Hollow Conducting Cylinder. What is the mag-
netic eld at points inside a hollow conducting cylinder which is made
such that its current density varies inversely as the distance from the cen-
ter of the cylinder? The conductor is shown in Fig. 11 and the current
density in this problem is:
j =
k
r
; a < r < b , (16)
where k is a constant. If a circular integration path is chosen, the current
enclosed by it, the right side of Eqs. (1) and (15), is:
I
C
=
_
r
a
k
r
2r
dr
B d
= 4k
m
I
C
.
It relates the integral of the magnetic eld around a closed loop to the
net current owing through any surface bounded by the integration
loop. Amperes law is universally true, but is useful only when there is
a high degree of symmetry.
current density: a vector whose magnitude at a space point is the
current per unit area normal to the direction of the current at that
point and whose direction is the direction of the current at that point.
line integral: the integral of a function along a specied path in
space. In Amperes law one evaluates the line integral of the tangential
component of the magnetic eld around a closed path that: (i) goes
through the point at which one wishes to know the magnetic eld; and
(ii) is such that it has a constant value for the integrand so the integral
can be performed trivially.
solenoid: a tightly wound cylindrical helix of current-carrying wire.
toroid: a solenoid bent into the shape of a doughnut.
14
MISN-0-138 11
A. Line Integrals
a
b
1
2
3 . . .
B
^
i
q
Dl
i
`
Figure 12.
The line integral
_
b
a
B d
,
for the path shown above, can be approximated by dividing the path into
many small segments
i
and for each segment the product
B
i
cos
i
i
can be found. Here B
i
cos
i
is the component of
B tangent to the curve.
The integral can be calculated approximately by summing these segments
terms, for example, on a computer. However, the exact value of the line
integral is given by the limit:
_
b
a
B d
= lim
n
i=1
B
i
cos
i
i
.
If a is joined to b, the path becomes closed and the resultant integral
_
C
B d
B d
=
_
b
a
B d = B
_
b
a
d = B
ab
,
15
MISN-0-138 12
where
ab
is the length of the path from a to b.
(ii)
B is always perpendicular to the path:
_
b
a
B d
= 0 .
B. Projection of an Area
If a planar (at) area S is projected onto another plane, the area A
on the projected-onto plane is given by:
A =
j nS . (19)
Here n is a unit vector normal to the plane
of the original area and
j is a unit vector
normal to the projected-onto plane (see
the sketch). This is entirely equivalent to
the statement that the areas are related by
the cosine of the angle between the planes
(again see the sketch):
j
^
n
^
S
A
q
A = S cos . (20)
By projection we mean that from every point on the periphery of
the original area S we drop a perpendicular to the projected-onto plane.
The locus of those points on the projected-onto plane dene the periphery
of the projected area A.
Equations (19)-(20) are easily proved by considering innitesimally-
wide straight line elements of the area A that are normal to the line of
intersection of the two planes. For each such element there is a projection
of it onto the projected-onto plane, and the areas of the two elements are
obviously related by the cosine of the angle between the planes. Since the
areas themselves are simply the integrals of the innitesimal areas, and
since the angle between the planes is independent of where one is in one of
the areas, the cosine can be pulled outside the integral and Eqs. (19)-(20)
are proved.
If the area S is curved (non-planar) then Eqs. (19)-(20) apply only
to innitesimal areas (which can be considered to be planar for these
16
MISN-0-138 13
purposes):
dA =
j ndS .
17
MISN-0-138 PS-1
PROBLEM SUPPLEMENT
Note: Problems 8, 9, and 10 also occur in this modules Model Exam.
1. Three innitely long parallel wires each carry a current I in the
direction shown below. What is
_
Bd for each of the three paths C
1
,
C
2
, and C
3
?
x
x
C
1
C
3
C
2
2. The magnetic eld in a certain region of space is given by
B = A
0
x x
where A
0
= 3 T/m, x is the x-coordinate of the point, and
x is a unit vector in the x-direction. In this region, con-
sider a rectangular path in the x-y plane whose sides are
parallel to the x and y axes respectively as shown below.
C
B A
D
y =3m
y =1m
x =1m x =5m
a. Evaluate the line integral of
B from A to B.
b. Do the same along the line from B to C.
18
MISN-0-138 PS-2
c. For C to D.
d. For D to A.
e. Evaluate the
_
B d around this closed path.
f. Determine the net current that must be crossing the x-y plane
through the rectangle ABCD.
3. A long cylindrical conductor of radius R has a uniform current density
j spread over its cross section. Determine the magnetic eld produced
at points r < R and r > R and sketch the magnitude of
B as a
function of r.
4. A very long non-conducting cylinder has N conduct-
ing wires placed tightly together around its circumfer-
ence and running parallel to its axis as shown below:
R
If each wire carries a current I, nd the magnetic eld at points
inside and outside the cylinder.
5.
a
b
A hollow cylindrical conductor of radii a and b has a current I uni-
formly spread over its cross section.
19
MISN-0-138 PS-3
a. Show that the magnetic eld inside the conductor (a < r < b) is:
B =
2k
m
I
_
r
2
a
2
_
(b
2
a
2
) r
b. Express B in terms of the current density j.
c. Show that when a 0 you get the same answer as in problem 3.
6. A long coaxial cable consists of two concentric conduc-
tors. The outside conductor carries a current I equal to
that in the inside conductor, but in the opposite direction.
a
c
b
Find the magnetic eld at these points:
a. inside the inner conductor (r < a),
b. between the conductors (a < r < b),
c. inside the outer conductor (b < r < c), and
d. outside the cable (r > c).
7.
Uniform current density j
`
directed out of the page
to
t
An innite, plane, conducting slab of thickness t carries a uniform
current density of j amperes per square meter directed out of the
page in the above diagram.
20
MISN-0-138 PS-4
a. Apply Amperes law to determine the magnetic eld at a height h
above the center line of the slab for h > t/2. Explain carefully how
you make use of symmetry in setting up your integration path.
b. Suppose your integration path had been a rectangular loop with
two sides parallel to the slab surface (as you must have used), but
with one parallel path a distance h above the center line and the
other a distance h
are greater
than t/2). Explain in this case, and without prior knowledge of
your nal answer, why Amperes law cannot tell you
B at points
h above the slab. Then show how the use of symmetry arguments
solves the problem.
c. Use the answer to part (a) and Amperes law to determine the
magnetic eld at points a distance y below the surface of the slab,
inside the material. What is the eld at the center line? Sketch
the direction of the eld at various points inside the slab.
8.
C
B A
D
3.0
y(m)
1.0
1.0 0
x(m)
5.0
B = 1.0 10
1
xteslas everywhere.
In a certain region of space, the magnetic eld intensity is uniform
and has the value of 10 teslas directed in the positive x-direction at
every point in the region. In this region consider a rectangular path
in the x-y plane from point A to point B parallel to the x-axis, B to
C parallel to the y-axis and D back to A parallel to the y-axis (see
the sketch above).
a. Evaluate the line integral of
B from A to B. [N]
b. Do the same for the line from B to C. [B]
c. Do the same for the line from C to D. [I]
21
MISN-0-138 PS-5
d. Do the same for the line from D to A. [P]
e. Evaluate the loop integral
_
B d for this closed path. Use the results
of parts (a)-(d) to nd your answer. [J]
f. From your answer to part (e), determine the net current that must
be crossing the x-y plane through this rectangle ABCD. [A]
9. Repeat Problem 8, parts (a) through (f) for the case where the mag-
netic eld in this region is now given by
B(x, y) = (A
0
+ A
1
y) x
where A
0
= 2.0 T, A
1
= 0.50 T/m and y is the y-coordinate of the
point.
a. [C]
b. [K]
c. [O]
d. [M]
e. [H]
f. [L]
10.
Path 2
Path 1
R
r
B
B
r
A
A
R = radius of the cylindrical conducting wire
j = the current per unit area (distributed uniformly) directed into
the page
r
A
= the distance from the center to point A outside the conductor
22
MISN-0-138 PS-6
r
B
= the distance from the center to point B inside the conductor.
Path 1 (solid line) is a circular path surrounding the cylinder concen-
tric with the cylindrical conductor passing through point A.
Path 2 (dashed line) is an arbitrary path surrounding the conductor,
also passing through point A.
a. What is
_
B d for each of the paths 1 and 2? [F]
b. Explain how symmetry enables you to evaluate
B at point A only
if you use path 1. [G]
Brief Answers:
1. Circular path: Net current = I
_
C1
B d
= 4k
m
I
Rectangular Path: Net Current = I + (I) = 0
_
C1
B d
= 0
Irregular Path: Net Current = I I I = I
_
C1
B d
= 4k
m
I
2. a. 36 mT
b. zero
c. 36 mT
d. zero
e. zero
f.
_
B d
x = 4k
m
jtx,
where B
= B
_
d = 2rB = 4k
m
jR
2
,
so at point A:
B = 2k
m
jR
2
r
H. 4.0 Tm
25
MISN-0-138 PS-9
I. 4.0 10
1
Tm Help: [S-4]
J.
_
B d
B d
= 0 ,
which implies B = 0 since the path is arbitrary.
S-2 (from TX-4e)
I =
_
S
j ndS with
j =
k
r
r, giving:
I =
_
2
0
_
r
a
_
k
r
_
r ( r)r
dr
d
= 2k
_
r
a
_
1
r
_
r
dr
= 2k
_
r
a
dr
= 2kr
|
r
a
= 2k(r a).
27
MISN-0-138 AS-2
S-3 (from TX-4e)
_
B d
= 4k
m
I
(B)(2r) = 8
2
k
m
k(r a)
B = 4k
m
k
_
r a
r
_
S-4 (from PS-problem 8)
An integral is just the limit of a sum:
_
B d
= lim
.
Note that
B
= xd
and
B = xB
so:
_
D
C
B x ( xd) = B
_
D
C
d = +B
_
D
C
dx = B
_
1.0 m
5.0 m
dx = 4.0B m.
28
MISN-0-138 ME-1
MODEL EXAM
1.
C
B A
D
3.0
y(m)
1.0
1.0 0
x(m)
5.0
B = 10 x teslas everywhere.
In a certain region of space, the magnetic eld intensity is uniform
and has the value of 10 teslas directed in the positive x-direction at
every point in the region. In this region consider a rectangular path in
the x-y plane from point A to point B parallel to the x-axis, B to C
parallel to the y-axis and D back to A parallel to the y-axis (see the
sketch above).
a. Evaluate the line integral of
B from A to B.
b. Do the same for the line from B to C.
c. Do the same for the line from C to D.
d. Do the same for the line from D to A.
e. Evaluate
_
B d around this closed path. Use the results of parts
(a)-(d) to nd your answer.
f. From your answer to part (e), determine the net current that must be
crossing the x-y plane through this rectangle ABCD.
2. Repeat Problem 1, parts (a) through (f) for the case where the mag-
netic eld in this region is given by
B = (A
0
+ A
1
y) x
where A
0
= 2 T, A
1
= 0.5 T/m and y is the y-coordinate of the point.
29
MISN-0-138 ME-2
3.
Path 2
Path 1
R
r
B
B
r
A
A
R = radius of the cylindrical conducting wire
j = the current per unit area (distributed uniformly) directed into the
page
r = the distance from the center to point A outside the conductor
r = the distance from the center to point B inside inside the conductor.
Path 1 (solid line) is a circular path surrounding the cylinder concentric
with the cylindrical conductor passing through point A.
Path 2 (dashed line) is an arbitrary path surrounding the conductor,
also passing through point A.
a. What is
_
B d for each of the paths 1 and 2?
b. Explain how symmetry enables you to evaluate
B at point A only
if you use path 1.
Brief Answers:
1. See Problem 8 in this modules Problem Supplement
2. See Problem 9 in this modules Problem Supplement
3. See Problem 10 in this modules Problem Supplement
30
31 32